2036046 - Crash in [@ java.lang.RuntimeException: at mozilla.components.browser.engine.gecko.GeckoEngine$$ExternalSyntheticLambda24.onShutdown(R8$$SyntheticClass:5)]

Status: NEW

(GeckoView :: General, defect)

Description

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

Nightly-only crash which has been reported with increasing frequency since Firefox for Android 150.0a1.

Crash report: https://crash-stats.mozilla.org/report/index/6285d6e8-f9a4-4c04-809e-c1f1b0260430

Top 10 frames:

0  mozilla.components.browser.engine.gecko.GeckoEngine$$ExternalSyntheticLambda24  onShutdown  R8$$SyntheticClass:5
1  org.mozilla.geckoview.GeckoRuntime$1  handleMessage  GeckoRuntime.java:31
2  org.mozilla.gecko.EventDispatcher$3  run  EventDispatcher.java:13
3  android.os.Handler  handleCallback  Handler.java:938
4  android.os.Handler  dispatchMessage  Handler.java:99
5  android.os.Looper  loopOnce  Looper.java:201
6  android.os.Looper  loop  Looper.java:288
7  android.app.ActivityThread  main  ActivityThread.java:7870
8  java.lang.reflect.Method  invoke  Method.java:-2
9  com.android.internal.os.RuntimeInit$MethodAndArgsCaller  run  RuntimeInit.java:548

Comment 1

[BugBot [:suhaib / :marco]]

The bug is linked to a topcrash signature, which matches the following criterion:

• Top 10 AArch64 and ARM crashes on nightly

For more information, please visit BugBot documentation.

Comment 2

[BugBot [:suhaib / :marco]]

The severity field is not set for this bug.
:owlish, could you have a look please?

For more information, please visit BugBot documentation.

Comment 3

[Donal Meehan [:dmeehan]]

This crash has been happening since 150.0, but it has not been investigated.
:mstange, I asked Claude to dig into it, and it found that Bug 2023004 might be the regressor here?

Comment 5

[Markus Stange [:mstange]]

These reports are parent process crashes.

This exception is thrown when the parent process GeckoThread dispatches a "Gecko:Exited" event, which it does when nativeRun terminates cleanly. This is an unexpected situation - on Android, we expect to remain in the Gecko event loop forever until the OS kills us (or until we crash).

Gecko exits the root nativeRun function in these cases (according to Claude):

• Normal XPCOM shutdown was driven to completion (something fired quit-application / called nsAppStartup::Quit, e.g. an AppShellService shutdown path, a profile-change scenario, or a programmatic quit).
• A Gecko-initiated restart (restarting = true in the bundle on line 512–516).

I don't currently see a way how the patches from bug 2023004 could have had an impact on this.

Comment 6

[Markus Stange [:mstange]]

Mr. Beedell, do you happen to still have the full log which contained the excerpt you posted in bug 2041158? There's a "XRE_main returned %d" log message we output here which, if present, would have been logged just before the Java exception information, and it might give us some insight into what happened. If there's no message, that could mean that the result code was zero.

Comment 7

[Mr Beedell, Roke Julian Lockhart [:B.RJL]]

The ring buffer has long since purged those logs, per github.com/F0x1d/LogFox/issues/251. I've not retained any duplicates of the logs from before the crash, and it has not reproduced since, on my cited environment.

Comment 8

[BugBot [:suhaib / :marco]]

Based on the topcrash criteria, the crash signatures linked to this bug are not in the topcrash signatures anymore.

For more information, please visit BugBot documentation.

Comment 9

[BugBot [:suhaib / :marco]]

The severity field is not set for this bug.
:owlish, could you have a look please?

For more information, please visit BugBot documentation.